Added AUTOHEAL_START_PERIOD to delay the first health check

README.md

@@ -24,12 +24,13 @@ b) Set ENV `AUTOHEAL_CONTAINER_LABEL=all` to watch all running containers.
 
 c) Set ENV `AUTOHEAL_CONTAINER_LABEL` to existing label name that has the value `true`.
 
-Note: You must apply `HEALTHCHECK` to your docker images first. See https://docs.docker.com/engine/reference/builder/#/healthcheck for details.
+Note: You must apply `HEALTHCHECK` to your docker images first. See https://docs.docker.com/engine/reference/builder/#healthcheck for details.
 
 ## ENV Defaults
 ```
 AUTOHEAL_CONTAINER_LABEL=autoheal
 AUTOHEAL_INTERVAL=5   # check every 5 seconds
+AUTOHEAL_START_PERIOD=0   # wait 0 second before first health check
 DOCKER_SOCK=/var/run/docker.sock
 ```
 

docker-entrypoint

@@ -33,7 +33,9 @@ if [ "$1" = 'autoheal' ] && [ -e ${DOCKER_SOCK} ]; then
     }
   fi
 
-  echo "Monitoring containers for unhealthy status"
+  echo "Monitoring containers for unhealthy status in ${AUTOHEAL_START_PERIOD} second(s)"
+  sleep ${AUTOHEAL_START_PERIOD:=0}
+
   while true; do
     sleep ${AUTOHEAL_INTERVAL:=5}